Tumor early screening process analysis method and system based on big data technology

Abstract

The present invention discloses a method and system for analyzing the process of early tumor screening based on big data technology, aiming to improve the efficiency and accuracy of early tumor screening. The method comprises: obtaining historical screening data of early tumor screening personnel, constructing a multi-hop association network based on knowledge graph technology; performing screening weight analysis based on the association network to determine the data acquisition weight; obtaining screening data based on the weight and performing tumor risk assessment; and optimizing the secondary screening process based on the assessment results. The system includes modules for data acquisition, knowledge graph construction, weight analysis, risk assessment, and process optimization. By combining big data with knowledge graphs, the present invention realizes intelligent and personalized screening processes, significantly improves screening efficiency and accuracy, and has broad application prospects.

Description

Technical Field

[0001] The present invention relates to the technical field of early tumor screening, and in particular to a method and system for analyzing the process of early tumor screening based on big data technology. Background Art

[0002] Early cancer screening is crucial for improving cancer patient survival and reducing treatment costs. However, traditional cancer screening methods still have limitations, such as incomplete data collection, single-method analysis, and a lack of intelligent optimization of the screening process, which compromise the accuracy and efficiency of screening results.

[0003] Currently, cancer screening primarily relies on biomarker testing, imaging scans, and pathology testing. However, the data correlations between different screening methods have not been fully explored, resulting in insufficient reliability of screening results. Furthermore, traditional screening processes often rely on fixed screening strategies that cannot be dynamically adjusted based on individual patient characteristics and historical screening data, making the screening process lack flexibility and targeting.

[0004] Knowledge graph technology can explore the correlations between different screening data, thereby optimizing the screening process; machine learning technology can train models based on a large amount of historical screening data to improve the accuracy of tumor risk assessment; data mining technology can discover potential tumor risk factors and improve screening efficiency.

[0005] Therefore, the research on the process analysis method and system of early cancer screening based on big data technology has important application value. In summary, there is an urgent need for a process analysis method and system of early cancer screening based on big data technology to overcome the shortcomings of traditional screening methods, improve the intelligent level of cancer screening, and achieve efficient and accurate early cancer screening. Summary of the Invention

[0006] In order to solve at least one of the above technical problems, the present invention proposes a method and system for analyzing the process of early tumor screening based on big data technology.

[0007] The first aspect of the present invention provides a method for analyzing the process of early cancer screening based on big data technology, comprising:

[0008] Obtain historical screening data of a preset number of early cancer screening participants, and construct a multi-hop association network between the historical screening data based on knowledge graph technology;

[0009] Performing screening weight analysis on the historical screening data according to the multi-hop association network, and determining a data acquisition weight for each screening data item according to the screening weight;

[0010] Acquiring screening data of a tumor early screening subject according to the data acquisition weight, and performing a tumor risk assessment on the tumor early screening subject according to the screening data to obtain a tumor risk assessment result;

[0011] The secondary screening process for early cancer screening personnel is optimized based on the tumor risk assessment results.

[0012] In this solution, the historical screening data of a preset number of cancer early screening personnel are obtained, and a multi-hop association network between the historical screening data is constructed based on the knowledge graph technology, specifically:

[0013] Obtaining historical screening data of a preset number of individuals undergoing early cancer screening, the historical screening data including biomarker detection data, image scan feature data, and pathological index data, and performing structured processing on the historical screening data to obtain structured screening data, the structured screening data including screening item type, screening time series, screening results, and medical record data;

[0014] Defining entity types and relationship types of the structured screening data based on knowledge graph technology, where entity types include personnel identifiers, screening items, biomarkers, imaging features, and pathological indicators; and relationship types include temporal dependencies between screening items, correlations between biomarkers and imaging features, and causal relationships between pathological indicators and screening results;

[0015] Building a graph database for the entity types and relationship types based on big data technology, importing the entity types and relationship types into the graph database, and building a knowledge graph network with screening items as nodes and multi-hop association paths as edges;

[0016] Traversing the knowledge graph network based on a breadth-first search algorithm to extract multiple association paths of broad entity types, wherein the multi-hop association paths include a first-hop association from a biomarker to an imaging feature, a second-hop association from an imaging feature to a pathological indicator, and a third-hop association of temporal dependencies across screening items;

[0017] Determining a data call relationship between each screening item based on the frequency of occurrence of the multi-hop association path, wherein the data call relationship includes automatically calling the screening item data corresponding to the second-hop association path when the first-hop association path is triggered, and adjusting the execution order of the screening items according to the third-hop association path;

[0018] The data call relationship is mapped into directed edge weights in a multi-hop association network to generate a multi-hop association network including screening items, screening item data nodes, call relationship edges and weight labels.

[0019] In this solution, the historical screening data is subjected to screening weight analysis based on the multi-hop association network, and the data acquisition weight of each screening data item is determined based on the screening weight, specifically:

[0020] Constructing a decision tree for each multi-hop path in the multi-hop association network according to the multi-hop association network, and determining the number of calls and path depth of the screening data item corresponding to each multi-hop path node in the early cancer screening process according to the decision tree;

[0021] Determining a data call frequency for each screening data item in the multi-hop association network according to the number of calls and the path depth;

[0022] The screening weight of each screening data item is determined according to the data call frequency, and the data acquisition weight of each screening data item is determined according to the screening weight.

[0023] In this solution, the screening data of the tumor early screening personnel are obtained according to the data acquisition weights, and the tumor risk assessment of the tumor early screening personnel is performed based on the screening data to obtain the tumor risk assessment results, which are specifically:

[0024] The screening data features of historical screening data were extracted based on the PCA algorithm, and a mapping relationship was formed between the screening data features and the corresponding pathological indicators to construct a screening data feature-pathological indicator data matrix;

[0025] Performing a tumor risk probability assessment on each pathological indicator data, integrating the tumor risk probability with the screening data feature-pathological indicator data matrix, and constructing a screening data feature-pathological indicator data-tumor risk probability matrix;

[0026] A gradient boosting tree algorithm is introduced to construct a tumor risk assessment model, wherein the screening data features-pathological index data-tumor risk probability matrix is ​​imported into the gradient boosting tree algorithm to construct a decision tree, and the tumor risk assessment model is trained according to the decision tree. The tumor risk assessment model also includes a feature extraction and fusion layer constructed based on the PCA algorithm;

[0027] Collecting screening data of early cancer screening personnel based on the data acquisition weights, the screening data including real-time biomarker detection data, image scanning feature data, and pathological index time series data;

[0028] Inputting the biomarker detection data and the image scanning feature data into the feature extraction and fusion layer of the tumor risk assessment model to perform cross-modal correlation analysis, extracting the correlation pattern between the biomarker concentration change and the image feature, generating a fusion feature vector, and constructing the fusion feature vector as the input set of the tumor risk assessment model;

[0029] The input set is imported into the tumor risk assessment model to perform tumor risk assessment to obtain a tumor risk assessment result.

[0030] In this solution, the input set is introduced into the tumor risk assessment model to perform tumor risk assessment to obtain a tumor risk assessment result, specifically:

[0031] Performing multi-dimensional feature decomposition based on the fused feature vector to extract a feature subset required for tumor risk probability calculation, and importing the feature subset into the tumor risk assessment model to calculate a tumor risk probability value;

[0032] When the tumor risk probability value is greater than a preset risk probability threshold, the biomarker features and image features with a fluctuation amplitude greater than the preset fluctuation threshold in the fused feature vector are extracted and calibrated as fluctuation features. The fluctuation features are then secondary verified and matched with the feature library of confirmed cases in historical screening data. When the matching degree is greater than the matching threshold, a high-risk warning signal is generated;

[0033] According to the high-risk warning signal, the decision tree splitting rule of the tumor risk assessment model is adjusted, the splitting weight of the screening data item corresponding to the fluctuation feature is increased to a preset priority, and the tumor risk probability value is recalculated. When the recalculated probability value is continuously greater than the preset risk probability threshold, a tumor risk assessment result including a high-risk feature marker is output;

[0034] When the recalculated probability value falls below the preset risk probability threshold, the current screening data and risk assessment process are imported into the manual review system, and the split weights of the screening data items corresponding to the fluctuation characteristics of the tumor risk assessment model are secondary corrected based on the manually fed back tumor risk probability results to generate the final tumor risk assessment results.

[0035] In this solution, the secondary screening process for early cancer screening personnel is optimized based on the tumor risk assessment results, specifically:

[0036] Obtaining a high-risk probability value and its associated screening items marked in the tumor risk assessment result, and when the high-risk probability value continuously exceeds a preset risk threshold, extracting screening items that have a third-hop association path with the high-risk probability value from the multi-hop association network as priority screening items;

[0037] Generate a screening item combination based on the execution sequence of the priority screening items, monitor the risk increase changes of the priority screening items in three consecutive assessment cycles in real time, and trigger a screening time compression strategy when it is detected that the risk increase of any priority screening item exceeds a preset increase threshold;

[0038] Based on the screening time compression strategy, the execution frequency of the screening items with risk increase exceeding the threshold is optimized, and the execution interval of the screening items is shortened to a preset minimum period. At the same time, the acquisition resolution of the image scanning feature data is adjusted according to the correlation between the image features of the second-hop association path in the multi-hop association network and the pathological indicators;

[0039] When the risk increase of the screening item after execution frequency optimization falls below the preset increase threshold in the subsequent two assessment cycles, the original screening item execution sequence is restored and secondary screening process data is generated.

[0040] A second aspect of the present invention further provides a tumor early screening process analysis system based on big data technology, the system comprising: a memory and a processor, the memory comprising a tumor early screening process analysis method program based on big data technology, and when the tumor early screening process analysis method program based on big data technology is executed by the processor, the following steps are implemented:

[0041] Obtain historical screening data of a preset number of early cancer screening participants, and construct a multi-hop association network between the historical screening data based on knowledge graph technology;

[0042] Performing screening weight analysis on the historical screening data according to the multi-hop association network, and determining a data acquisition weight for each screening data item according to the screening weight;

[0043] Acquiring screening data of a tumor early screening subject according to the data acquisition weight, and performing a tumor risk assessment on the tumor early screening subject according to the screening data to obtain a tumor risk assessment result;

[0044] The secondary screening process for early cancer screening personnel is optimized based on the tumor risk assessment results.

[0045] The present invention discloses a method and system for analyzing the process of early tumor screening based on big data technology, aiming to improve the efficiency and accuracy of early tumor screening. The method comprises: obtaining historical screening data of early tumor screening personnel, constructing a multi-hop association network based on knowledge graph technology; performing screening weight analysis based on the association network to determine the data acquisition weight; obtaining screening data based on the weight and performing tumor risk assessment; and optimizing the secondary screening process based on the assessment results. The system includes modules for data acquisition, knowledge graph construction, weight analysis, risk assessment, and process optimization. By combining big data with knowledge graphs, the present invention realizes intelligent and personalized screening processes, significantly improves screening efficiency and accuracy, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of a method for analyzing the process of early cancer screening based on big data technology is shown in the present invention;

[0047] Figure 2 A flow chart showing the present invention's determination of a data acquisition weight for each screening data item;

[0048] Figure 3 A flow chart showing the present invention's optimization of the secondary screening process for early cancer screening personnel;

[0049] Figure 4 A block diagram of a tumor early screening process analysis system based on big data technology of the present invention is shown. DETAILED DESCRIPTION

[0050] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0052] Figure 1 A flowchart of a tumor early screening process analysis method based on big data technology of the present invention is shown.

[0053] like Figure 1 As shown, the first aspect of the present invention provides a method for analyzing the process of early cancer screening based on big data technology, comprising:

[0054] S102, obtaining historical screening data of a preset number of cancer early screening individuals, and constructing a multi-hop association network among the historical screening data based on knowledge graph technology;

[0055] S104, performing screening weight analysis on the historical screening data according to the multi-hop association network, and determining a data acquisition weight for each screening data item according to the screening weight;

[0056] S106, obtaining screening data of the cancer early screening subject according to the data acquisition weight, performing a tumor risk assessment on the cancer early screening subject according to the screening data, and obtaining a tumor risk assessment result;

[0057] S108, optimizing the secondary screening process for early cancer screening personnel based on the tumor risk assessment results.

[0058] It should be noted that by acquiring historical screening data from a predetermined number of individuals undergoing early cancer screening and constructing a multi-hop association network based on knowledge graph technology, this system enables structured storage and deep association mining of screening data, improving the usability and analytical value of screening data. The multi-hop association network is used to analyze screening weights on historical screening data and, based on these weights, determine the data acquisition weights for each screening data item. This optimizes data retrieval strategies, reduces redundant data collection, and improves the accuracy and utilization of screening data. Based on these optimized weights, screening data from individuals undergoing early cancer screening is collected and combined with biomarker data, image scan feature data, and pathology data for cancer risk assessment. PCA feature extraction and a gradient boosting tree model are used to improve the accuracy and robustness of risk assessment, enabling personalized risk prediction. Furthermore, based on the cancer risk assessment results, the secondary screening process for high-risk individuals is optimized. By analyzing screening items associated with high-risk features in the multi-hop association network, the order and time intervals of screening items are adjusted, and image scan resolution and screening frequency are dynamically optimized to improve the targeting and efficiency of screening. Ultimately, this results in more accurate and efficient early cancer screening, increasing early detection rates and screening effectiveness.

[0059] According to an embodiment of the present invention, the historical screening data of a preset number of early cancer screening personnel are obtained, and a multi-hop association network between the historical screening data is constructed based on the knowledge graph technology, specifically:

[0060] Obtaining historical screening data of a preset number of individuals undergoing early cancer screening, the historical screening data including biomarker detection data, image scan feature data, and pathological index data, and performing structured processing on the historical screening data to obtain structured screening data, the structured screening data including screening item type, screening time series, screening results, and medical record data;

[0061] Defining entity types and relationship types of the structured screening data based on knowledge graph technology, where entity types include personnel identifiers, screening items, biomarkers, imaging features, and pathological indicators; and relationship types include temporal dependencies between screening items, correlations between biomarkers and imaging features, and causal relationships between pathological indicators and screening results;

[0062] Building a graph database for the entity types and relationship types based on big data technology, importing the entity types and relationship types into the graph database, and building a knowledge graph network with screening items as nodes and multi-hop association paths as edges;

[0063] Traversing the knowledge graph network based on a breadth-first search algorithm to extract multiple association paths of broad entity types, wherein the multi-hop association paths include a first-hop association from a biomarker to an imaging feature, a second-hop association from an imaging feature to a pathological indicator, and a third-hop association of temporal dependencies across screening items;

[0064] Determining a data call relationship between each screening item based on the frequency of occurrence of the multi-hop association path, wherein the data call relationship includes automatically calling the screening item data corresponding to the second-hop association path when the first-hop association path is triggered, and adjusting the execution order of the screening items according to the third-hop association path;

[0065] The data call relationship is mapped into directed edge weights in a multi-hop association network to generate a multi-hop association network including screening items, screening item data nodes, call relationship edges and weight labels.

[0066] It's important to note that building a multi-hop association network enables efficient organization of cancer screening data, intelligent association analysis, and optimized screening processes, thereby improving screening accuracy and resource utilization. A multi-hop association network is a data association structure constructed based on knowledge graph technology. Nodes are composed of screening items, biomarkers, imaging features, and pathological indicators, while directed edges represent temporal dependencies between screening items, associations between biomarkers and imaging features, and causal relationships between pathological indicators and screening results. This creates a multi-layered network structure. The multi-hop nature of this network is reflected in the fact that the associations between data nodes are not simple one-to-one direct connections, but rather information is transmitted through multiple physical nodes, enabling logical reasoning and information expansion of screening data. The first hop association (biomarker-imaging feature): Abnormal levels of certain biomarkers (such as alpha-fetoprotein (AFP) and carcinoembryonic antigen (CEA)) may indicate a patient's potential risk of malignancy. Changes in these biomarkers are often correlated with imaging features (such as tumor volume changes and enhancement patterns in MRI or CT scans). Therefore, when biomarker data is abnormal, the system can automatically trigger an imaging scan, improving the specificity of imaging diagnosis and reducing unnecessary imaging tests. Second-hop association (imaging feature-pathology marker): Imaging studies may reveal suspicious tumor features, such as unclear borders, vascular proliferation, and abnormal enhancement patterns. However, imaging results alone cannot fully confirm the malignancy of a tumor. Therefore, based on the abnormal imaging features, the system automatically associates and matches the most relevant pathology tests, such as biopsy histology and immunohistochemistry, to ensure that pathology tests are prioritized for high-risk patients and avoid unnecessary invasive tests. Third-hop association (temporal dependency across screening tests): Different screening tests may have temporal dependencies. For example, a patient's low-dose CT (LDCT) screening result six months ago was normal, but recent biomarker data showed significantly elevated alpha-fetoprotein. Based on this temporal dependency, the system can identify the changing trend between the patient's past normal screening results and the current abnormal biomarker data, automatically adjusting the screening plan and pre-arranging high-precision imaging tests (such as PET-CT) or circulating tumor DNA (ctDNA) testing to improve the timeliness of risk assessment. These big data technologies include building distributed computing databases and data warehouses. Graph databases are used to store and manage tumor screening data and its complex relationships. They utilize a graph data structure composed of nodes, edges, and attributes, enabling efficient storage and query of entities such as screening items, biomarkers, imaging features, and pathological indicators, as well as their relationships. Compared to traditional relational databases (such as MySQL and PostgreSQL), graph databases are more suitable for processing medical screening data with complex relationships and multi-hop connections.

[0067] Figure 2 A flow chart of determining the data acquisition weight of each screening data item according to the present invention is shown.

[0068] According to an embodiment of the present invention, the screening weight analysis is performed on the historical screening data according to the multi-hop association network, and the data acquisition weight of each screening data item is determined according to the screening weight, specifically:

[0069] S202, constructing a decision tree for each multi-hop path in the multi-hop association network based on the multi-hop association network, and determining the number of calls and path depth of the screening data item corresponding to each multi-hop path node in the early cancer screening process based on the decision tree;

[0070] S204, determining a data call frequency of each screening data item in the multi-hop association network according to the call count and the path depth;

[0071] S206 , determining a screening weight for each screening data item according to the data call frequency, and determining a data acquisition weight for each screening data item according to the screening weight.

[0072] It is important to note that by calculating the frequency of data retrieval for different screening data items across multi-hop pathways, high-impact and highly relevant screening data can be identified, thereby improving screening data utilization and reducing redundant data retrieval. For example, when a biomarker (such as AFP) is abnormal, the system can intelligently match its associated imaging scans (such as MRI) and pathological indicators (such as Ki-67 immunohistochemistry) to prioritize the execution of necessary screening items. Secondly, the system dynamically adjusts data acquisition strategies based on the screening weights of different screening items, improving the flexibility and efficiency of the screening process. For high-risk individuals, the system automatically increases the data acquisition weights for key screening items (such as ctDNA analysis and PET-CT) to enhance early diagnosis capabilities. For low-risk individuals, unnecessary imaging scans or pathological tests are reduced, with only periodic monitoring of key biomarkers performed, thereby reducing screening costs and patient burden. Furthermore, based on the real-time calculation of screening weights, this method adaptively optimizes the execution order of screening items, avoiding ineffective screenings, ensuring a more accurate and efficient screening process and improving the success rate of early cancer screening.

[0073] According to an embodiment of the present invention, the screening data of the tumor early screening personnel is obtained according to the data acquisition weight, and the tumor risk assessment is performed on the tumor early screening personnel based on the screening data to obtain the tumor risk assessment result, which is specifically:

[0074] The screening data features of historical screening data were extracted based on the PCA algorithm, and a mapping relationship was formed between the screening data features and the corresponding pathological indicators to construct a screening data feature-pathological indicator data matrix;

[0075] Performing a tumor risk probability assessment on each pathological indicator data, integrating the tumor risk probability with the screening data feature-pathological indicator data matrix, and constructing a screening data feature-pathological indicator data-tumor risk probability matrix;

[0076] A gradient boosting tree algorithm is introduced to construct a tumor risk assessment model, wherein the screening data features-pathological index data-tumor risk probability matrix is ​​imported into the gradient boosting tree algorithm to construct a decision tree, and the tumor risk assessment model is trained according to the decision tree. The tumor risk assessment model also includes a feature extraction and fusion layer constructed based on the PCA algorithm;

[0077] Collecting screening data of early cancer screening personnel based on the data acquisition weights, the screening data including real-time biomarker detection data, image scanning feature data, and pathological index time series data;

[0078] Inputting the biomarker detection data and the image scanning feature data into the feature extraction and fusion layer of the tumor risk assessment model to perform cross-modal correlation analysis, extracting the correlation pattern between the biomarker concentration change and the image feature, generating a fusion feature vector, and constructing the fusion feature vector as the input set of the tumor risk assessment model;

[0079] The input set is imported into the tumor risk assessment model to perform tumor risk assessment to obtain a tumor risk assessment result.

[0080] It is important to note that feature extraction from historical screening data using the PCA algorithm effectively reduces dimensionality and removes redundant information, identifying key features strongly correlated with pathological indicators. This creates a screening data feature-pathological indicator matrix. This matrix maps and correlates multidimensional screening data (e.g., biomarker concentrations and image scan texture features) with pathological results (e.g., histological grade and degree of malignancy), providing structured input for subsequent model training. A tumor risk assessment model constructed by incorporating the gradient boosting tree algorithm adaptively learns the nonlinear relationship between different screening features and tumor risk. It optimizes feature weights layer by layer using the splitting rule of a decision tree, thereby capturing the complex correlation patterns between dynamic changes in biomarkers and abnormal imaging features. During the real-time assessment phase, the model performs cross-modal correlation analysis through a feature extraction and fusion layer (a neural network module that integrates biomarker test data with image scan feature data). For example, it jointly models the mutation frequency of circulating tumor DNA (ctDNA) in serum with the tumor volume growth rate in MRI images to generate a fused feature vector (a numerical representation that comprehensively reflects the correlation between multidimensional screening indicators). This enables the model to identify early-stage tumor risk signals that are difficult to detect using single-modality data. In addition, the model can optimize the risk assessment path for specific screening data of different patients by dynamically adjusting the splitting priority of feature subsets. For example, when a patient's imaging features show suspicious lesions but the biomarkers are not significantly abnormal, the model will automatically enhance the weight of the imaging features in the decision tree and make probability corrections based on the pathological results of similar cases in historical data to avoid missed diagnoses.

[0081] According to an embodiment of the present invention, the step of importing the input set into the tumor risk assessment model to perform tumor risk assessment and obtain a tumor risk assessment result is specifically as follows:

[0082] Performing multi-dimensional feature decomposition based on the fused feature vector to extract a feature subset required for tumor risk probability calculation, and importing the feature subset into the tumor risk assessment model to calculate a tumor risk probability value;

[0083] When the tumor risk probability value is greater than a preset risk probability threshold, the biomarker features and image features with a fluctuation amplitude greater than the preset fluctuation threshold in the fused feature vector are extracted and calibrated as fluctuation features. The fluctuation features are then secondary verified and matched with the feature library of confirmed cases in historical screening data. When the matching degree is greater than the matching threshold, a high-risk warning signal is generated;

[0084] It is important to note that due to individual variability and physiological fluctuations in biomarkers (such as ctDNA concentration) and imaging features (such as tumor margin enhancement), relying solely on risk probability thresholds can easily lead to false positives (e.g., misinterpreting an inflammatory-induced AFP elevation as a risk factor for liver cancer) or missed detections (e.g., early-stage tumors with ambiguous imaging features but biomarker abnormalities). For example, if a patient experiences a transient elevation in alpha-fetoprotein (AFP) due to short-term liver function abnormalities, triggering an alert based solely on the risk probability threshold could lead to unnecessary invasive testing. Meanwhile, in another patient, although the overall risk probability does not reach the threshold, the number of circulating tumor cells (CTCs) increases stepwise and PET-CT metabolic values ​​are slightly abnormal. This co-fluctuation pattern may be overlooked by traditional models. Therefore, by extracting cross-modal features with fluctuation amplitudes exceeding the threshold (e.g., a sudden 50% increase in the coefficient of variation of a biomarker accompanied by increased image texture heterogeneity) and performing pattern matching with a historical case database (e.g., comparing the AFP fluctuation curve of a liver cancer patient in the three months prior to diagnosis with the enhancement characteristics of the arterial phase on enhanced CT), we can effectively distinguish physiological fluctuations from malignant lesions. When a patient's fusion features match 80% of the pre-diagnosis feature trajectories in the historical liver cancer case library, the system will generate a high-risk warning even if the initial risk probability value only slightly exceeds the threshold, thereby resolving the clinical pain point that "early signals of occult tumors are easily diluted by the overall probability" and improving the ability to capture progressive malignant changes.

[0085] According to the high-risk warning signal, the decision tree splitting rule of the tumor risk assessment model is adjusted, the splitting weight of the screening data item corresponding to the fluctuation feature is increased to a preset priority, and the tumor risk probability value is recalculated. When the recalculated probability value is continuously greater than the preset risk probability threshold, a tumor risk assessment result including a high-risk feature marker is output;

[0086] It's important to note that statically assigned weights may not dynamically capture cross-modal fluctuations in a specific patient's screening data that are strongly associated with malignancy. For example, if a patient's ctDNA mutation abundance increases by 50% and MRI shows accelerated nodule enhancement, such coordinated fluctuations may indicate early malignancy. However, if the model defaults to prioritizing imaging features (e.g., as secondary split nodes), this risk may be underestimated. By increasing the split weights of fluctuating features (e.g., ctDNA mutations and MRI enhancement rate) to a preset priority (e.g., increasing the Gini coefficient gain weight of the node split from third to first), the model prioritizes these features in the data, strengthening their risk contribution. Recalculating the probability verifies the persistence of feature abnormalities: if the adjusted risk value remains above the threshold, it indicates that the fluctuating feature is not sporadic noise (e.g., inflammatory interference) but rather is associated with stable tumor progression. This mechanism overcomes the traditional model's lack of sensitivity to sudden malignant signals, ensuring accurate early warning of feature abnormalities, avoiding the risk of missed diagnoses due to parameter rigidity, and improving the ability to capture progressive malignancy in dynamic screening.

[0087] When the recalculated probability value falls below the preset risk probability threshold, the current screening data and risk assessment process are imported into the manual review system, and the split weights of the screening data items corresponding to the fluctuation characteristics of the tumor risk assessment model are secondary corrected based on the manually fed back tumor risk probability results to generate the final tumor risk assessment results.

[0088] It's important to note that when the risk probability falls below the threshold after model weight adjustment, the initial fluctuation may have been due to incidental interference (e.g., detection error or short-term physiological abnormality). For example, a patient's CA19-9 biomarker suddenly increases, accompanied by changes in pancreatic morphology on CT scans. After weight adjustment, the risk value briefly exceeds the threshold, but due to the patient's history of chronic pancreatitis, the value naturally decreases upon reexamination. Directly accepting the model's results in this situation could lead to misjudgment due to overreliance on the algorithm (mislabeling inflammatory lesions as pancreatic cancer risk) and miss specific malignancy features that the model hasn't learned (e.g., the specific fluctuation patterns of rare neuroendocrine tumors). Manual review incorporates clinical judgment. When a clinician confirms that a fluctuation feature is unrelated to malignancy, the system reduces the split weight of the associated feature (e.g., reducing the weight decay coefficient of the CA19-9-pancreatic morphology association path from 0.8 to 0.3) to prevent overreaction to similar interference data. If a clinician discovers a malignancy sign not identified by the model (e.g., a specific gene mutation combined with an imaging calcification pattern), the weight of the new feature combination is reversed. This dynamic correction mechanism avoids the accumulation of false positives caused by the algorithm's "overconfidence".

[0089] Figure 3 A flow chart showing the present invention's optimization of the secondary screening process for early cancer screening personnel is shown.

[0090] According to an embodiment of the present invention, the secondary screening process for early cancer screening personnel is optimized based on the tumor risk assessment results, specifically as follows:

[0091] S302, obtaining a high-risk probability value and its associated screening items marked in the tumor risk assessment result, and when the high-risk probability value continuously exceeds a preset risk threshold, extracting screening items that have a third-hop association path with the high-risk probability value from the multi-hop association network as priority screening items;

[0092] S304: Generate a screening item combination based on the execution sequence of the priority screening items, monitor the risk increase changes of the priority screening items in real time over three consecutive assessment cycles, and trigger a screening time compression strategy when it is detected that the risk increase of any priority screening item exceeds a preset increase threshold;

[0093] S306, optimizing the execution frequency of the screening items for which the risk increase exceeds the threshold value based on the screening time compression strategy, shortening the execution interval of the screening items to a preset minimum period, and adjusting the acquisition resolution of the image scan feature data based on the correlation between the image features of the second-hop association path in the multi-hop association network and the pathological index;

[0094] S308, when the risk increase of the screening item after execution frequency optimization falls below the preset increase threshold in the subsequent two evaluation cycles, the original screening item execution sequence is restored and secondary screening process data is generated.

[0095] It is important to note that the extraction of third-hop associated screening items based on a multi-hop association network (for example, automatically associating a patient's abnormally elevated circulating tumor cell count with previously unnoticed micronodule features from a low-dose CT scan six months prior) can deeply uncover hidden risk associations across time, ensuring that prioritized screening items accurately target the early biological behavior of potentially malignant lesions, avoiding the blindness of traditional fixed screening programs. Secondly, by monitoring risk increases in real time and triggering screening time compression strategies (for example, shortening the MRI review interval from six months to three months), the frequency of tracking high-risk features is dynamically increased. Combined with adaptive image resolution adjustment, this optimizes medical resource allocation while maintaining screening sensitivity, avoiding over-screening of low-risk patients. Finally, a mechanism that restores the original timing when risk increases decreases (for example, discontinuing high-frequency biopsies after risk indicators stabilize) enables flexible management of the screening process, preventing the physical and mental burden of long-term, high-frequency screening on patients. Furthermore, the continuous generation of secondary screening process data provides closed-loop feedback for model iteration, thereby improving the overall clinical applicability of the cancer early screening system and its ability to intercept malignant lesions at an early stage. The correlation between imaging features and pathological indicators refers to the strength of the statistical or clinical correlation between the lesion characteristics (such as tumor morphology and enhancement pattern) observed by medical imaging technology (such as CT and MRI) and the pathological examination results (such as histological typing and degree of malignancy), which is obtained through correlation analysis through logistic regression.

[0096] According to an embodiment of the present invention, the further embodiment includes:

[0097] Obtaining tumor risk assessment result data for the early cancer screening population, determining a screening urgency score for the secondary screening process for each early cancer screening individual based on the tumor risk assessment result data, and constructing a dynamic priority queue based on the screening urgency score;

[0098] Obtaining real-time schedule data for the hospital's early cancer screening equipment and doctor scheduling data, and establishing a resource availability matrix based on the real-time schedule data and doctor scheduling data, including equipment idle periods, doctor workload, and the time consumption of early screening examination items;

[0099] A dual-objective optimization model of screening urgency and resource constraints is constructed based on a linear programming algorithm. The dynamic priority queue and resource availability matrix are imported into the dual-objective optimization model. The dynamic priority queue and resource availability matrix input in real time are matched based on an online reinforcement learning algorithm to determine the examination time for each early screening item for each early cancer screening person, thereby obtaining a cancer early screening equipment allocation plan.

[0100] The execution status of the assigned inspection task equipment in the early cancer screening equipment allocation plan and the new high-risk inspection case data of the hospital are monitored in real time. When a critical case is detected and inserted into the dynamic priority queue, the insertion position of the priority queue is determined according to the criticality of the case to obtain a resource adjustment plan for the early cancer screening equipment.

[0101] It should be noted that in the field of early cancer screening, existing medical resource scheduling systems generally use static scheduling rules, which make it difficult to dynamically respond to the contradiction between the changing urgency of high-risk population screening and the real-time fluctuations in medical resources. This leads to problems such as long waiting times for high-risk cases, low equipment utilization, and uneven workload among radiologists. The present invention achieves efficient dynamic allocation of screening resources by constructing a dynamic priority queue driven by screening urgency scores, combining multidimensional constraint modeling of the resource availability matrix, and a dual-objective optimization model based on linear programming and reinforcement learning. This solution can quickly respond to the screening needs of high-risk cases and significantly shorten the waiting time for examinations. At the same time, through real-time monitoring of equipment status and a critical case insertion mechanism, it ensures that critical cases have priority access to high-precision imaging equipment resources, improving equipment utilization and response speed. In addition, the resource availability matrix integrates multidimensional data such as equipment idle time, physician load, and examination time, and combines it with the online learning capabilities of reinforcement learning to optimize the workload distribution of radiologists, reduce the occupancy rate of low-risk cases on key resources, and achieve efficient resource utilization. The present invention significantly improves hospital screening throughput while ensuring screening timeliness.

[0102] Figure 4 A block diagram of a tumor early screening process analysis system based on big data technology of the present invention is shown.

[0103] A second aspect of the present invention further provides a tumor early screening process analysis system 4 based on big data technology. The system includes: a memory 41 and a processor 42. The memory includes a tumor early screening process analysis method program based on big data technology. When the tumor early screening process analysis method program based on big data technology is executed by the processor, the following steps are implemented:

[0104] Obtain historical screening data of a preset number of early cancer screening participants, and construct a multi-hop association network between the historical screening data based on knowledge graph technology;

[0105] Performing screening weight analysis on the historical screening data according to the multi-hop association network, and determining a data acquisition weight for each screening data item according to the screening weight;

[0106] Acquiring screening data of a tumor early screening subject according to the data acquisition weight, and performing a tumor risk assessment on the tumor early screening subject according to the screening data to obtain a tumor risk assessment result;

[0107] The secondary screening process for early cancer screening personnel is optimized based on the tumor risk assessment results.

[0108] The present invention discloses a method and system for analyzing the process of early tumor screening based on big data technology, aiming to improve the efficiency and accuracy of early tumor screening. The method comprises: obtaining historical screening data of early tumor screening personnel, constructing a multi-hop association network based on knowledge graph technology; performing screening weight analysis based on the association network to determine the data acquisition weight; obtaining screening data based on the weight and performing tumor risk assessment; and optimizing the secondary screening process based on the assessment results. The system includes modules for data acquisition, knowledge graph construction, weight analysis, risk assessment, and process optimization. By combining big data with knowledge graphs, the present invention realizes intelligent and personalized screening processes, significantly improves screening efficiency and accuracy, and has broad application prospects.

[0109] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0110] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0111] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0112] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0113] Alternatively, if the integrated units described above are implemented as software modules and sold or used as standalone products, they can also be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product, stored on a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute all or part of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as removable storage devices, ROM, RAM, magnetic disks, or optical disks.

[0114] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for analyzing the process of early cancer screening based on big data technology, characterized in that: The following steps are involved: Obtain historical screening data of a preset number of early cancer screening participants, and construct a multi-hop association network between the historical screening data based on knowledge graph technology; Performing screening weight analysis on the historical screening data according to the multi-hop association network, and determining a data acquisition weight for each screening data item according to the screening weight; Acquiring screening data of a tumor early screening person according to the data acquisition weight, and performing a tumor risk assessment on the tumor early screening person according to the screening data of the tumor early screening person to obtain a tumor risk assessment result; Optimizing the secondary screening process for early cancer screening personnel based on the tumor risk assessment results; The method of obtaining historical screening data of a preset number of early cancer screening personnel and constructing a multi-hop association network among the historical screening data based on the knowledge graph technology is as follows: Obtaining historical screening data of a preset number of individuals undergoing early cancer screening, the historical screening data including biomarker detection data, image scan feature data, and pathological index data, and performing structured processing on the historical screening data to obtain structured screening data, the structured screening data including screening item type, screening time series, screening results, and medical record data; Defining the entity types and relationship types of the structured screening data based on knowledge graph technology, wherein entity types include personnel identification and screening items, and relationship types include temporal dependencies between screening items, correlations between biomarkers and imaging features, and causal relationships between pathological indicators and screening results. The screening items include biomarkers, imaging features, and pathological indicators. Building a graph database for the entity types and relationship types based on big data technology, importing the entity types and relationship types into the graph database, and building a knowledge graph network with screening items as nodes and multi-hop association paths as edges; Traversing the knowledge graph network based on a breadth-first search algorithm to extract a multi-hop association path of entity types, wherein the multi-hop association path includes a first-hop association from a biomarker to an imaging feature, a second-hop association from an imaging feature to a pathological indicator, and a third-hop association of temporal dependencies across screening items; Determining a data call relationship between each screening item based on the frequency of occurrence of the multi-hop association path, wherein the data call relationship includes automatically calling the screening item data corresponding to the second-hop association path when the first-hop association path is triggered, and adjusting the execution order of the screening items according to the third-hop association path; The data call relationship is mapped into directed edge weights in a multi-hop association network to generate a multi-hop association network including screening items, screening item data nodes, call relationship edges and weight labels.

2. The method for analyzing tumor early screening process based on big data technology according to claim 1, characterized in that: The screening weight analysis is performed on the historical screening data according to the multi-hop association network, and the data acquisition weight of each screening data item is determined according to the screening weight, specifically: Constructing a decision tree for each multi-hop path in the multi-hop association network according to the multi-hop association network, and determining the number of calls and path depth of the screening data item corresponding to each multi-hop path node in the early cancer screening process according to the decision tree; Determining a data call frequency for each screening data item in the multi-hop association network according to the number of calls and the path depth; The screening weight of each screening data item is determined according to the data call frequency, and the data acquisition weight of each screening data item is determined according to the screening weight.

3. The method for analyzing tumor early screening process based on big data technology according to claim 1, characterized in that: The screening data of the tumor early screening personnel are obtained according to the data acquisition weights, and the tumor risk assessment is performed on the tumor early screening personnel according to the screening data to obtain the tumor risk assessment results, which are specifically: The screening data features of historical screening data were extracted based on the PCA algorithm, and a mapping relationship was formed between the screening data features and the corresponding pathological indicators to construct a screening data feature-pathological indicator data matrix; Performing a tumor risk probability assessment on each pathological indicator data, integrating the tumor risk probability with the screening data feature-pathological indicator data matrix, and constructing a screening data feature-pathological indicator data-tumor risk probability matrix; A gradient boosting tree algorithm is introduced to construct a tumor risk assessment model, wherein the screening data features-pathological index data-tumor risk probability matrix is ​​imported into the gradient boosting tree algorithm to construct a decision tree, and the tumor risk assessment model is trained according to the decision tree. The tumor risk assessment model also includes a feature extraction and fusion layer constructed based on the PCA algorithm; Collecting screening data of early cancer screening personnel based on the data acquisition weights, the screening data including real-time biomarker detection data, image scanning feature data, and pathological index time series data; Inputting the biomarker detection data and the image scanning feature data into the feature extraction and fusion layer of the tumor risk assessment model to perform cross-modal correlation analysis, extracting the correlation pattern between the biomarker concentration change and the image feature, generating a fusion feature vector, and constructing the fusion feature vector as the input set of the tumor risk assessment model; The input set is imported into the tumor risk assessment model to perform tumor risk assessment to obtain a tumor risk assessment result.

4. The method for analyzing tumor early screening process based on big data technology according to claim 3, characterized in that: The step of importing the input set into the tumor risk assessment model to perform tumor risk assessment and obtain a tumor risk assessment result is specifically as follows: Performing multi-dimensional feature decomposition based on the fused feature vector to extract a feature subset required for tumor risk probability calculation, and importing the feature subset into the tumor risk assessment model to calculate a tumor risk probability value; When the tumor risk probability value is greater than a preset risk probability threshold, the biomarker features and image features with a fluctuation amplitude greater than the preset fluctuation threshold in the fused feature vector are extracted and calibrated as fluctuation features. The fluctuation features are then secondary verified and matched with the feature library of confirmed cases in historical screening data. When the matching degree is greater than the matching threshold, a high-risk warning signal is generated; According to the high-risk warning signal, the decision tree splitting rule of the tumor risk assessment model is adjusted, the splitting weight of the screening data item corresponding to the fluctuation feature is increased to a preset priority, and the tumor risk probability value is recalculated. When the recalculated probability value is continuously greater than the preset risk probability threshold, a tumor risk assessment result including a high-risk feature marker is output; When the recalculated probability value falls below the preset risk probability threshold, the current screening data and risk assessment process are imported into the manual review system, and the split weights of the screening data items corresponding to the fluctuation characteristics of the tumor risk assessment model are secondary corrected based on the manually fed back tumor risk probability results to generate the final tumor risk assessment results.

5. The method for analyzing tumor early screening process based on big data technology according to claim 1, characterized in that: The secondary screening process for early cancer screening personnel is optimized based on the tumor risk assessment results, specifically: Obtaining a high-risk probability value and its associated screening items marked in the tumor risk assessment result, and when the high-risk probability value continuously exceeds a preset risk threshold, extracting screening items that have a third-hop association path with the high-risk probability value from the multi-hop association network as priority screening items; Generate a screening item combination based on the execution sequence of the priority screening items, monitor the risk increase changes of the priority screening items in three consecutive assessment cycles in real time, and trigger a screening time compression strategy when it is detected that the risk increase of any priority screening item exceeds a preset increase threshold; Based on the screening time compression strategy, the execution frequency of the screening items with risk increase exceeding the threshold is optimized, and the execution interval of the screening items is shortened to a preset minimum period. At the same time, the acquisition resolution of the image scanning feature data is adjusted according to the correlation between the image features of the second-hop association path in the multi-hop association network and the pathological indicators; When the risk increase of the screening item after execution frequency optimization falls below the preset increase threshold in the subsequent two assessment cycles, the original screening item execution sequence is restored and secondary screening process data is generated.

6. A tumor early screening process analysis system based on big data technology, characterized in that: The tumor early screening process analysis system based on big data technology includes a storage device and a processor. The storage device includes a tumor early screening process analysis method program based on big data technology. When the tumor early screening process analysis method program based on big data technology is executed by the processor, the following steps are implemented: Obtain historical screening data of a preset number of early cancer screening participants, and construct a multi-hop association network between the historical screening data based on knowledge graph technology; Performing screening weight analysis on the historical screening data according to the multi-hop association network, and determining a data acquisition weight for each screening data item according to the screening weight; Acquiring screening data of a tumor early screening person according to the data acquisition weight, and performing a tumor risk assessment on the tumor early screening person according to the screening data of the tumor early screening person to obtain a tumor risk assessment result; Optimizing the secondary screening process for early cancer screening personnel based on the tumor risk assessment results; The method of obtaining historical screening data of a preset number of early cancer screening personnel and constructing a multi-hop association network among the historical screening data based on the knowledge graph technology is as follows: Obtaining historical screening data of a preset number of individuals undergoing early cancer screening, the historical screening data including biomarker detection data, image scan feature data, and pathological index data, and performing structured processing on the historical screening data to obtain structured screening data, the structured screening data including screening item type, screening time series, screening results, and medical record data; Defining the entity types and relationship types of the structured screening data based on knowledge graph technology, wherein entity types include personnel identification and screening items, and relationship types include temporal dependencies between screening items, correlations between biomarkers and imaging features, and causal relationships between pathological indicators and screening results. The screening items include biomarkers, imaging features, and pathological indicators. Building a graph database for the entity types and relationship types based on big data technology, importing the entity types and relationship types into the graph database, and building a knowledge graph network with screening items as nodes and multi-hop association paths as edges; Traversing the knowledge graph network based on a breadth-first search algorithm to extract a multi-hop association path of entity types, wherein the multi-hop association path includes a first-hop association from a biomarker to an imaging feature, a second-hop association from an imaging feature to a pathological indicator, and a third-hop association of temporal dependencies across screening items; Determining a data call relationship between each screening item based on the frequency of occurrence of the multi-hop association path, wherein the data call relationship includes automatically calling the screening item data corresponding to the second-hop association path when the first-hop association path is triggered, and adjusting the execution order of the screening items according to the third-hop association path; The data call relationship is mapped into directed edge weights in a multi-hop association network to generate a multi-hop association network including screening items, screening item data nodes, call relationship edges and weight labels.

Images